Ride-sharing and on-demand services are becoming more common in cities to complement existing public transportation offerings. The Rhein-Neckar-Verkehr GmbH (RNV) in Mannheim is also considering introducing demand-based mobility services.
Where and when does the use of demand-based mobility services make sense? With the help of an expert opinion, PTV consultants identified possible deployment areas and times. Especially important here was that future offerings should complement existing public transport rather than cannibalizing it, especially with regard to urban rail.
The PTV team determined how high the potential demand for on-demand transport is in various areas. Using simulations, parameters of the offerings were defined that allow for sensible and efficient operation.
Selection of the operation areas
A PTV Visum transport model was deployed to analyze possible areas of operation and times. In addition, PTV’s consultants evaluated data from the RNV's automatic passenger counting system. They evaluated passenger flows from past passenger surveys and motorized private transport volumes from transport development planning data.
All evaluations were then visualized. Using this analysis, it was possible to formulate a proposal for the intelligent division of operation areas.
Transport model as basis
For the simulation of on-demand transport, first the experts adapted an existing transport model. New virtual stops were inserted and these were connected to the transport zones. This model was used to make KPI calculations that serve as the basis for the simulation.
Furthermore, the consultants estimated what demand could be anticipated and what the traffic patterns of on-demand transport might look like. In the process, a distinction was made between local and long-distance trips and the effects of shifting on individual and public transport were determined.
Simulation and route planning with PTV technology
Various issues were examined in the simulation with PTV software: How many vehicles and what capacities are required for operational efficiency? How can vehicle use be minimized and the number of requested trips served maximized? What is the situation with regard to the range of electric vehicles and their charge status? How long are passengers willing to wait for the autonomous on-demand shuttle? This way, various service and operating parameters such as average travel time, average detour factor, fulfillment quota, occupancy rate, idle kilometers, transport performance, stop time, as well as other operational and transport-related KPIs could be examined.
In Mannheim, on-demand transport makes particular sense in the outer areas of the city to complement existing public transport. This way, times and places can be covered that are currently less well-served.
In the long term, demand-based services can help redesign the bus system to provide better overall coverage. On the other hand, routes (tangential connections) that are currently slow and involve many transfers can be made more attractive.
The simulation provided the following insights:
- Occupancy rates of 1.5 and higher are realistic
- Significant trip pooling will be achieved
- Significant additional traffic can arise due to these trips
- A high quality and service level will be achieved
- For efficient operation, a certain area size or a larger collection of areas is required.
The results with regard to the intelligent creation of areas and schedules for on-demand transport and its transit effects will be used for the implementation process, among other things for purchasing vehicles, positioning charging stations, personnel planning, and the approval of virtual stops.